able to put on a spurt when they encounter an adverse current. If a machine is made that is able to maintain itself in the air for any considerable length of time, it will not be a very difficult task to know when a current of air of this kind is encountered, because, if the engine is working up to speed, and everything is in perfect order, and still the machine is falling, it is very certain that an unfavourable current has been encountered, and efforts should be made to get out of it as soon as possible. Then, again, if the machine has an abnormal tendency to rise without any increase in the number of rotations made by the screws, the aeronaut may be certain that he has encountered an upward and favourable current of air which, unfortunately, will not last. It should, however, be borne in mind that, while the width of the upward current is not very great, nevertheless, it may extend in a practically straight line for many miles. CHAPTER IV. PRINCIPALLY RELATING TO SCREWS. In 1887 I was approached by several wealthy gentlemen who asked me if I thought it was possible to make a flying machine. I said, “certainly; the domestic goose is able to fly and why should not man be able to do as well as a goose?” They then asked me what it would cost and how long it would take, and, without a moment’s hesitation, I said it would require my undivided attention for five years and might cost £100,000. A great deal of experimenting would be necessary; the first three years would be devoted to developing an internal combustion engine of the Brayton or Otto type, and the next two years to experimenting with aeroplanes and screws and building a machine. Even at that time I had a clear idea of the system that would be the best. However, nothing was then done, but in 1889 I employed for the purpose two very skilful American mechanics, and put them to work at Baldwyn’s Park, Kent. At that time the petroleum motor had not been reduced to its present degree of efficiency and lightness; it was not suitable for a flying machine, and I saw that it would require a lot of experimental work in order to develop it. After taking into consideration all the facts of the case, I decided to use a steam engine. Had I been able to obtain the light and efficient motors which have been recently developed, thanks to the builders of racing cars, I should not have had to experiment at all with engines and boilers, as I could have obtained the necessary motors at once. As it was, I was obliged to content myself with the steam engine. [Illustration: Fig. 9.--Group of screws and other objects used in my experiments.] [Illustration: Fig. 10.--Some of the principal screws experimented with--_h_, a screw with very thick blades, and _g_, a screw made after a French model.]